Search

Decades before its clinical onset, epigenetic changes start to accumulate in the progenitor cells of Acute Myelogenous Leukemia (AML). Delineating these changes can improve risk-stratification for patients and shed insights into AML etiology, dynamics and mechanisms. Towards this goal, we extracted “epigenetic signatures” through two parallel machine learning approaches: a supervised regression model using frequently mutated genes as labels and an unsupervised topic modeling approach to factorize covarying epigenetic changes into a small number of “topics”. First, we created regression models forDNMT3AandTET2, the two most frequently mutated epigenetic drivers in AML. Our model differentiated wild-type vs. mutant genotypes based on their downstream epigenetic impacts with very high accuracy: AUROC 0.9 and 0.8, respectively. Methylation loci frequently selected by the models recapitulated known downstream pathways and identified several novel recurrent targets. Second, we used topic modeling to systematically factorize the high dimensional methylation profiles to a latent space of 15 topics. We annotated identified topics with biological and clinical features such as mutation status, prior malignancy and ELN criteria. Topic modeling successfully deconvoluted the combined effects of multiple upstream epigenetic drivers into individual topics including relatively infrequent cytogenetic events, improving the methylation-based subtyping of AML. Furthermore, they revealed complimentary and synergistic interactions between drivers, grouped them based on the similarity of their downstream methylation impact and linked them to prognostic criteria. Our models identify new signatures and methylation pathways, refine risk-stratification and inform detection and drug response studies for AML patients.KEY POINTSSupervised and unsupervised models reveal new methylation pathways of AML driver events and validate previously known associations.IndividualDNMT3AandTET2signatures are accurate and robust, despite the complex genetic and epigenetic make-up of samples at diagnosis.Unsupervised topic modeling factorizes covarying methylation changes and isolates methylation signatures caused by rare mutations.Topic modeling reveals a group of mutations with similar downstream methylation impacts and mapped to adverse-risk class by ELN.Topic modeling uncovers methylation signatures of infrequent cytogenetic events, significantly improving methylation-based subtyping.Our models can be leveraged to build predictive models for AML-risk.Our models show that cytogenetic events, such as t(15;17) have widespreadtransdownstream methylation impacts.

Decades before its clinical onset, epigenetic changes start to accumulate in the progenitor cells of Acute Myelogenous Leukemia– caused by tumor intrinsic driver mutations and tumor extrinsic environmental factors. There are millions of cancer-associated epigenetic changes that are persistent and stable throughout the leukemogenesis and at relapse. By grouping correlated loci into “epigenetic signatures”, we can gain insights on the AML etiology, dynamics as well as mechanisms around the methylation pathways of driver events. We approached this problem in two parts: First, we computationally constructed and analyzed methylation signatures that are associated with DNMT3A and TET2 mutations. Both of them are frequently mutated in AML and they are major causes of changes in the methylation landscape of healthy individuals, conferring a broad risk of transformation to a myeloid malignancy with the serial acquisition of an additional cooperating mutation. We built highly accurate ridge regression models that can differentiate the wild type vs mutant genotypes for DNMT3A and TET2 based on their downstream epigenetic impacts (AUROC 0.9 and 0.8, respectively). It suggests that despite the complicated genetic and epigenetic make-up of post-diagnosis samples, epigenetic regulator mutations’ signatures are distinguishable, and these distinctive marks remain stable over time. We further evaluated methylation loci selected by the models and showed that known downstream targets of DNMT3A such as EVI1/MECOM and AEBP2 (polycomb repressive complex member) were frequently selected by the models. We also identified multiple novel recurrent sites that were previously unassociated with DNMT3A and TET2. Second, we modeled different sources of variations in epigenetic patterns with Topic Modeling. Topic Modeling maps the sparse and high dimensional space of methylation profiles to a small and non-sparse latent space of small number of factors (topics). Each sample is represented as a combination of few topics in that latent space. We annotated these topics with biological and clinical features such as mutation status, prior malignancy, age and gender to shed light on the prevalence and combination of the mutations and other extrinsic factors present at the time of diagnosis. In the future, we aim to identify the etiologies of new signatures to inform detection, monitoring, risk stratifying and drug response studies for AML patients and track these changes throughout leukemogenesis in early detection cohorts. Citation Format: Burcu Gurun, Jeffrey W. Tyner, Emek Demir, Brian J. Druker, Paul T. Spellman. Computational modeling of methylation impact of AML driver events [abstract]. In: Proceedings of the AACR Special Conference: Acute Myeloid Leukemia and Myelodysplastic Syndrome; 2023 Jan 23-25; Austin, TX. Philadelphia (PA): AACR; Blood Cancer Discov 2023;4(3_Suppl):Abstract nr A15.